Rising players board
Ascension: who's rising
Every NHL skater's last few games against his previous 20: more ice time, a PP1 spot, a promotion to the top six, minutes with his team's best players and better chances. Each player gets a 0–100 score with the reasons spelled out, so you can grab the riser before your league notices, and see who is losing his role.
2026-27 · games through Oct 20, 2026 · 638 skaters scored on the last 5 games · updated Oct 21, 2026.
Biggest risers
- 5 P in last 5 (0.30/GP before)
- PP1 in 5 of last 5 (was 1 of 20)
- +1:26 PP TOI
- 5 P in last 5 (0.30/GP before)
- +1:06 PP TOI
- 20% of team PP time (was 0%)
- ixG/60 2.35 (was 0.80)
- +2:54 TOI a game
- Top-six in 2 of last 5 (was 2 of 20)
- ixG/60 0.99 (was 0.61)
- +2:31 PP TOI
- PP1 in 5 of last 5 (was 5 of 20)
- On-ice xGF/60 4.21 (was 2.39)
- +4:25 TOI a game
- 74% of 5v5 TOI with Hughes (was 6%)
Biggest fallers: losing their role
- On-ice xGF/60 1.13 (was 2.78)
- −2:09 PP TOI
- PP1 in 0 of last 5 (was 15 of 20)
- 0 P in last 5 (0.85/GP before)
- PP1 in 3 of last 5 (was 18 of 20)
- Line Watch: moved down a line Oct 15
- On-ice xGF/60 2.16 (was 3.77)
- PP1 in 2 of last 5 (was 20 of 20)
- −0:58 PP TOI
- On-ice xGF/60 2.16 (was 3.65)
- −1:29 TOI a game
- 43% of 5v5 TOI with top players (was 100%)
- 3 P in last 5 (1.10/GP before)
- −2:23 PP TOI
- PP1 in 2 of last 5 (was 17 of 20)
The board
| # | Player | Score | Tier | Wk | Rostered | Why |
|---|---|---|---|---|---|---|
| 1 | 11% rostered | 86 | Surging | ▲ 16 | 11% |
|
| 2 | Filling in for Jordan Staal on L2 (+1 more out) (IR, calf, est. return Dec 3 · ESPN) 23% rostered | 85 | Surging | ±0 | 23% |
|
| 3 | Filling in for Rasmus Dahlin on D2/PP1 (+1 more out) (no timeline · scratch report) 32% rostered | 85 | Surging | ±0 | 32% |
|
| 4 | Filling in for Jared Spurgeon on D1 (+1 more out) (IR, upper body, est. return Oct 27 · ESPN) 8% rostered | 83 | Surging | ▲ 19 | 8% |
|
| 5 | New team · 8 GP with NSH (from NYR) · filling in for Mavrik Bourque on L1 (day-to-day, upper body · ESPN) 23% rostered | 83 | Surging | ±0 | 23% |
|
| 6 | Filling in for Marco Rossi on L3/PP1 (day-to-day, illness · ESPN) 27% rostered | 83 | Surging | ▲ 9 | 27% |
|
| 7 | New team · 8 GP with EDM (from LAK) · filling in for Ryan Nugent-Hopkins on L2/PP1 (IR, lower body, est. return Oct 22 · ESPN) 4% rostered | 82 | Surging | ▲ 8 | 4% |
|
| 8 | Promoted to L1/PP1 for the last 8 games (was L2) 13% rostered | 82 | Surging | ▼ 1 | 13% |
|
| 9 | 7% rostered | 82 | Surging | ▲ 7 | 7% |
|
| 10 | New team · 9 GP with SJS (from BUF) 7% rostered | 81 | Surging | ▲ 37 | 7% |
|
| 11 | New team · 9 GP with FLA (from OTT) 5% rostered | 81 | Surging | ▲ 4 | 5% |
|
| 12 | Filling in for Taylor Hall on L1 (+2 more out) (IR, upper body, est. return Nov 7 · ESPN) 17% rostered | 81 | Surging | ▼ 3 | 17% |
|
| 13 | 21% rostered | 81 | Surging | ▲ 2 | 21% |
|
| 14 | New team · 8 GP with BUF (from ANA) · filling in for Rasmus Dahlin on D2/PP1 (no timeline · scratch report) 33% rostered | 81 | Surging | ▲ 8 | 33% |
|
| 15 | Promoted to L1/PP1 for the last 8 games (was L4/PP2) 42% rostered | 81 | Surging | ▼ 3 | 42% |
|
| 16 | Promoted to D1 for the last 4 games (was D2) 44% rostered | 80 | Surging | ▼ 1 | 44% |
|
| 17 | Promoted to L1/PP1 for the last 4 games (was L1/PP2) 45% rostered | 79 | Surging | ▲ 3 | 45% |
|
| 18 | Filling in for Samuel Honzek on L2 (IR, lower body, est. return Oct 28 · ESPN) 12% rostered | 78 | Surging | ▲ 2 | 12% |
|
| 19 | Promoted to D1 for the last 4 games (was D2) 9% rostered | 78 | Surging | ▲ 21 | 9% |
|
| 20 | Sean MonahanC · CBJ 16% rostered | 77 | Surging | ▼ 6 | 16% |
|
| 21 | Cam FowlerD · STL 15% rostered | 77 | Surging | ▲ 6 | 15% |
|
| 22 | Filling in for Matt Duchene on L2/PP1 (LTIR, abdomen, est. return Jan 12 · ESPN) 9% rostered | 77 | Surging | ▲ 14 | 9% |
|
| 23 | Filling in for Samuel Honzek on L2 (IR, lower body, est. return Oct 28 · ESPN) 11% rostered | 77 | Surging | ▲ 6 | 11% |
|
| 24 | 31% rostered | 77 | Surging | ▲ 1 | 31% |
|
| 25 | Luke Evangelista is back on L1/PP1; he lost the slot 24% rostered | 77 | Surging | ▲ 22 | 24% |
|
| 26 | Tye KartyeL · NYR 16% rostered | 77 | Surging | ▼ 4 | 16% |
|
| 27 | Pavel Buchnevich is back on L2; he lost the slot 5% rostered | 76 | Surging | ▲ 22 | 5% |
|
| 28 | Promoted to D1 for the last 5 games (was D3) 35% rostered | 75 | Surging | ▲ 3 | 35% |
|
| 29 | New team · 8 GP with CGY (from NJD) 48% rostered | 75 | Surging | ▲ 18 | 48% |
|
| 30 | Filling in for Troy Terry on L2/PP1 (LTIR, hip, est. return Nov 21 · ESPN) 5% rostered | 75 | Surging | ▲ 15 | 5% |
|
| 31 | Danila YurovR · MIN 37% rostered | 74 | Rising | ▲ 4 | 37% |
|
| 32 | Promoted to L3/PP1 for the last 2 games (was L2/PP2) 31% rostered | 74 | Rising | ▲ 2 | 31% |
|
| 33 | 0% rostered | 74 | Rising | ▲ 11 | 0% |
|
| 34 | New team · 5 GP with PIT (from EDM) · filling in for Trevor van Riemsdyk on D1 (IR, lower body, est. return Oct 22 · ESPN) 0% rostered | 73 | Rising | ▲ 25 | 0% |
|
| 35 | Filling in for Calum Ritchie on L1 (+1 more out) (out, lower body, est. return Dec 1 · ESPN) 0% rostered | 73 | Rising | ▲ 2 | 0% |
|
| 36 | Filling in for Troy Terry on L2/PP1 (LTIR, hip, est. return Nov 21 · ESPN) 6% rostered | 73 | Rising | ▲ 10 | 6% |
|
| 37 | Emmitt FinnieC · DET 31% rostered | 73 | Rising | ▼ 6 | 31% |
|
| 38 | Filling in for Matias Maccelli on L4/PP1 (+2 more out) (out, lower body, est. return Nov 17 · ESPN) 14% rostered | 72 | Rising | ▲ 58 | 14% |
|
| 39 | Promoted to D2 for the last 2 games (was D3) 5% rostered | 72 | Rising | ▲ 20 | 5% |
|
| 40 | Ryker EvansD · SEA 18% rostered | 72 | Rising | ▲ 4 | 18% |
|
| 41 | Filling in for Mike Matheson on D1 (day-to-day, lower body · ESPN) 15% rostered | 72 | Rising | ▲ 17 | 15% |
|
| 42 | New team · 10 GP with TOR (from TBL) · filling in for John Tavares on L2/PP1 (IR, lower body, est. return Nov 27 · ESPN) 10% rostered | 72 | Rising | ▲ 2 | 10% |
|
| 43 | 30% rostered | 71 | Rising | ▲ 22 | 30% |
|
| 44 | New team · 8 GP with NJD (from FLA) · filling in for Luke Evangelista on L1/PP1 (inferred from lineup) 8% rostered | 71 | Rising | ▲ 38 | 8% |
|
| 45 | New team · 8 GP with NSH (from UTA) · filling in for Mavrik Bourque on L1 (day-to-day, upper body · ESPN) 3% rostered | 71 | Rising | ▼ 3 | 3% |
|
| 46 | Filling in for Jared Spurgeon on D1 (IR, upper body, est. return Oct 27 · ESPN) 0% rostered | 70 | Rising | ▲ 25 | 0% |
|
| 47 | New team · 9 GP with MIN (from NYI) 33% rostered | 70 | Rising | ▼ 18 | 33% |
|
| 48 | New team · 6 GP with BOS (from PIT) · filling in for Charlie McAvoy on D1/PP1 (inferred from lineup) 6% rostered | 70 | Rising | ±0 | 6% |
|
| 49 | Filling in for Jordan Staal on L2 (+1 more out) (IR, calf, est. return Dec 3 · ESPN) 7% rostered | 70 | Rising | ▼ 10 | 7% |
|
| 50 | 4% rostered | 70 | Rising | ▼ 2 | 4% |
|
New role and want to know what it led to for similar players? See Breakout Comps.
How Ascension works
We take each skater's last 3, 5 or 10 games (you pick the window) and compare them with his previous 20 games. Early in the season the 20 reach back into last season. Six components measure the change:
- Ice time
- All-situations, 5v5 and power-play TOI per game in the window against his previous 20 games. A two-minute jump in 5v5 time or a minute more on the power play counts for a lot; a one-game spike is shrunk away.
- Power-play role
- His share of the team's PP time and how many games he spent on PP1 (detected from shift data), compared with before. "PP1 in 4 of last 5 (was 0 of 10)" is the classic riser.
- Line role
- His line slot each game from our line detector: L1 counts 1, L2 0.67, L3 0.33, L4 0 (D1 1, D2 0.5, D3 0). Moving into the top six or top four shows up here.
- Elite linemates
- The share of his 5v5 ice time spent with his team's top three scorers (points per game over their last 20 games), from shared-TOI pairs. "61% of 5v5 TOI with McDavid" is a role, not luck.
- Underlying production
- Individual expected goals per 60 (our xG v1 model), on-ice 5v5 xGF/60 and points per game, window vs baseline. It carries the most weight because it was the strongest predictor in the backtest.
- Line Watch events
- Promotions and demotions our deployment tracker flagged in the window (PP1 in or out, line up or down, ice-time jumps or drops), weighted by their significance score.
Each component is a signed value from −1 to +1 (the tanh of the change over a typical move), shrunk by sample size: the window counts for n / (n + 1.5) and the baseline for b / (b + 3). The weights are ice time 0.15, power-play role 0.15, line role 0.05, elite linemates 0.05, underlying production 0.55, line watch events 0.05. The score is 50 + 50 × tanh(Σ weight × value / 0.4): 50 means no change. 75+ is Surging, 60+ Rising, 41–59 Steady, 26–40 Cooling and 25 or lower Falling.
Full details, including how the weights were fit: Ascension methodology.
Fill-ins, new roles and promotions
Every player on the board can carry a label that says why his role changed. Labels are display only: the score stays the same.
- New role
- He changed teams and has played at least one game for the new club. His role there is still settling, so these players get their own tab.
- Fill-in
- A regular is out (injured, scratched or off the roster) and he took that player's line or power-play slot, or at least a minute more a game at the same position. The reason line names the player he is covering for and, when ESPN has one, the expected return date.
- Next up
- A regular is out and he is next in line for the spot, but he has not clearly taken it yet.
- Coach promotion
- Moved up a line or onto PP1 for at least two straight games with nobody out: the coach chose him.
- Small sample
- Fewer than 3 games in the window or 10 in the baseline. The score is shrunk toward 50, but read it with care.
- Production only
- Almost all of his score is points and chances, with no more ice time or a better role behind it.
- Role ending
- The regular he was covering for is back in the lineup and he lost the slot.
Fill-ins also get a durability bucket built from facts, not a model: Short (the regular is day-to-day or due back within 7 days), Medium (8–14 days), Long (more than 14), Unknown (out with no reported return) and Lasting (a new team, or the regular is off the roster). Injury reports come from ESPN, checked every hour and every 20 minutes on game nights; lineup scratches are read after each game, so a regular who sits two straight games without a report shows as out with no timeline.
The Value sort: tonight, this week, rest of season
Sort by Value to rank the same players by what they should add to your team over the horizon you pick: Tonight, This week or Rest of season. Each player's per-game stat line is his rest-of-season projection (LineupExperts, with our derived model filling power-play points, +/- and faceoffs), multiplied out over his team's games in the horizon and his own availability (nothing while he is out, half a game tonight if he is day-to-day). The totals are rated like the Week Planner: z-scores against the top 300 skaters for category leagues (+/- counts half on Tonight and This week; hits, PIM and faceoff wins are capped at +3), or projected fantasy points for points leagues. We also tested adding a fill-in's or promoted player's extra minutes on top of his projection: over two seasons it made the next-game and next-four-game forecasts slightly worse in every category, so it is switched off until a better role model wins.
Does it work? The backtest
We scored every skater every week of 2024-25 and 2025-26 and checked his points per game over the next 10 games. The weights were fit on 2025-26, Oct 27 – Dec 29 (6,020 player-weeks); the rows below are games the fit never saw.
| Holdout | Player-weeks | R² baseline → with score | Error cut | Surging next 10 | Falling next 10 |
|---|---|---|---|---|---|
| 2025-26, Jan – Mar | 6,788 | 0.487 → 0.501 | 1.3% | +0.08 P/GP, 58% beat baseline | −0.09 P/GP, 32% beat baseline |
| 2024-25, Nov – Mar (no pair data) | 10,761 | 0.426 → 0.440 | 1.2% | +0.10 P/GP, 61% beat baseline | −0.10 P/GP, 29% beat baseline |
The honest caveat: over just 10 games, a simple "previous 20 plus last 5 points" model predicts points as well as the score does (R² 0.515). Ascension's value is that it sees the role change (PP1, the top line, more minutes) before or without the points, and tells you why.
FAQ
What is the Ascension score?
A 0–100 score of how much a skater's role and underlying production have changed over his last 3, 5 or 10 games compared with his previous 20. 50 means no change; Surging (75+) players are gaining ice time, power-play time, a better line or better linemates and are producing more chances.
Does a high Ascension score predict more points?
A little, and we publish how much. On games the weights never saw (2025-26, Jan – Mar), Surging players averaged +0.08 points per game more over their next 10 games than in their previous 20, and 58% beat their baseline; Falling players averaged -0.09 and only 32% beat it. Adding the score to a baseline-only model cut the error by 1.3%. Ten-game point totals are noisy, so treat it as a tiebreaker and an early warning, not a guarantee.
Is this just a hot streak list?
No. Points are only one part of one component. A player can be Surging with zero points if he has moved onto PP1 and the top line with more ice time, which is exactly when to pick him up. Check the Luck Meter as well: a riser who is also running hot is more likely to cool off.
Why is a player I know is rising not on the board?
Players need at least five games before the window as a baseline, and anyone who has not played in three weeks drops off. Early in the season the baseline reaches back into last season, which the board notes. The Regulars only filter also hides depth players under 12 minutes a night.
How often does the board update?
After each night's games: the board refreshes when the night's games are ingested and again in the morning once Line Watch events are re-scored.
Can I see only players I can actually pick up?
Yes. By default the board shows players available in an average league: rostered in 50% of leagues or fewer (LineupExperts rostered %), and you can switch the cut to 25% or 75% or show everyone. Connect your league and you can show only the players who are free in it.
What do Fill-in, New role and Coach promotion mean?
They say why his role changed. A fill-in took the slot of a regular who is out, and the board names that player and his expected return. New role means he changed teams. A coach promotion is a move up the lineup with nobody out. Labels never change the score; they tell you how long the bump is likely to last.
What is the Value sort?
A second ranking that turns each player's role into fantasy value for your scoring: the Yahoo standard, roto, banger, classic or points presets, your own custom weights, or your connected league's settings. It counts the games his team plays in the horizon (back-to-backs are marked) and whether he is available for them. Signed in, your choices are saved to your account.
Why is Rest of season marked Beta?
Rest of season is his season projection over his team's remaining games. It does not yet account for how long a new role lasts, because our role model has not passed its backtest on two seasons of role changes. Treat it as a rough guide; Tonight and This week depend mostly on the schedule.